System and method for predictive maintenance of a battery assembly using temporal signal processing

• US 8,154,297 B2
• Filed: 07/05/2007
• Issued: 04/10/2012
• Est. Priority Date: 07/05/2007
• Status: Active Grant

• Current Assignee: ABB Schweiz AG (ABB Limited)
• Original Assignee: ABB Technology AG (ABB Limited)
• Inventors: Mousavi, Mirrasoul J., Maharsi, Mohamed, Bayoumi, Deia, McGowan, John J.
• Primary Examiner(s): Tso, Edward
• Assistant Examiner(s): PIGGUSH, AARON C
• Application Number:
  US11/773,504
• Publication Number:
  US 20090009183A1
• Time in Patent Office: 1,741 Days
• Field of Search: 324/426, 324/429, 324/430, 324/433, 324/537, 320/134, 320/136, 320/157
• US Class Current: 324/426
• CPC Class Codes: G01R 31/3648 comprising digital calculation means, e.g. for performing an algorithm
• CPC Class Codes: G01R 31/367 Software therefor, e.g. for battery testing using modelling or look-up tables
• CPC Class Codes: G05B 23/0232 based on qualitative trend analysis, e.g. system evolution
• CPC Class Codes: G05B 23/027 Alarm generation, e.g. communication protocol; Forms of alarm

Abstract

A technique of monitoring a battery assembly may include monitoring a parameter associated with the battery assembly to obtain a number of monitored parameter samples. A temporal sequence of monotonically increasing values may be generated from the monitored parameter samples. The temporal sequence may be analyzed for an indication of a trend in the monitored parameter toward one of an upper operational boundary or a lower operational boundary to predict a fault condition of the battery assembly.

Claims

1. A method of monitoring a battery assembly, comprising:

monitoring a parameter associated with the battery assembly to obtain a number of monitored parameter samples;

generating a temporal sequence of monotonically increasing values from the monitored parameter samples, the sequence including at least three values;

predicting a fault condition of the battery assembly when the temporal sequence of monotonically increasing values indicates a sustained inclination in the monitored parameter toward one of an upper predetermined value or a lower predetermined value, the upper and lower predetermined values selected to correspond to a condition for which the battery is monitored; and

upon making the prediction, triggering an output to a user.

2. The method of claim 1, wherein the predicting includes deriving an alarm signal from the temporal sequence, the alarm signal having a series of values based on an application of a Laplace test statistic to the temporal sequence values.

3. The method of claim 2, wherein the predicting further includes comparing the alarm signal against at least one of an upper threshold or a lower threshold.

4. The method of claim 1, the predicting including:

generating an alarm signal that indicates a deviation in the temporal sequence of monotonically increasing values toward an upper threshold set to correspond to the upper predetermined value or a lower threshold set to correspond to the lower predetermined value;

comparing the alarm signal to the upper threshold and to the lower threshold; and

making the prediction if the alarm signal violates the upper threshold or the lower threshold for a predetermined number of alarm signal values.

5. The method of claim 2, wherein the predicting further includes normalizing the alarm signal.

6. The method of claim 5, wherein the predicting further includes comparing the normalized alarm signal against at least one of an upper threshold or a lower threshold.

7. The method of claim 6, further comprising generating an alarm if the normalized alarm signal violates the upper threshold or the lower threshold for a predetermined number of alarm signal values.

8. The method of claim 6, further comprising generating an alarm if the normalized alarm signal violates the upper threshold or the lower threshold.

9. The method of claim 1, wherein the battery assembly is at least part of a lead acid battery system.

10. The method of claim 1, wherein the battery assembly is monitored while in use.

11. The method of claim 10, wherein the battery assembly is a primary power source for a load.

12. The method of claim 10, wherein the battery assembly is a backup power source for a load.

13. The method of claim 1, wherein the battery assembly is used in an electrical power transmission and distribution substation.

14. The method of claim 1, wherein the monitored parameter is a function of at least one of voltage, current, impedance or temperature.

15. The method of claim 4, further comprising:

comparing the alarm signal to another upper threshold that indicates less deviation in the temporal sequence of monotonically increasing values than the upper threshold and to another lower threshold that indicates less deviation in the temporal sequence of monotonically increasing values than the lower threshold, the another upper and lower thresholds corresponding to a smaller degree of the monitored condition than the upper and lower thresholds; and

if the alarm signal violates the another upper threshold or the another lower threshold for a predetermined number of alarm signal values, triggering an output to a user indicative of the smaller degree of the monitored condition.

16. A non-transitory computer readable medium storing a program, the program for monitoring a battery assembly and comprising:

code that generates a temporal sequence of monotonically increasing values based on a series of monitored parameter samples, the monitored parameter samples obtained by monitoring a parameter associated with the battery assembly, the sequence including at least three values; and

code that predicts a fault condition of the battery assembly when the temporal sequence of monotonically increasing values indicates a sustained inclination in the monitored parameter toward one of an upper predetermined value or a lower predetermined value, the upper and lower predetermined values selected to correspond to a battery condition for which the battery is monitored; and

code that triggers an output to a user upon making the prediction.

17. The non-transitory computer readable medium of claim 16, wherein the code that predicts derives an alarm signal from the temporal sequence, the alarm signal having a series of values based on an application of a Laplace test statistic to the temporal sequence values.

18. The non-transitory computer readable medium of claim 16, wherein the code that predicts includes code that:

generates an alarm signal that indicates a deviation in the temporal sequence of monotonically increasing values toward an upper threshold set to correspond to the upper predetermined value or a lower threshold set to correspond to the lower predetermined value;

compares the alarm signal to the upper threshold and the lower threshold; and

makes the prediction if the alarm signal violates the upper threshold or the lower threshold for a predetermined number of alarm signal values.

19. The non-transitory computer readable medium of claim 18, wherein the representation of the alarm signal is a normalized alarm signal.

20. The non-transitory computer readable medium of claim 18, further comprising code that:

compares the alarm signal to another upper threshold that indicates less deviation in the temporal sequence of monotonically increasing values than the upper threshold and to another lower threshold that indicates less deviation in the temporal sequence of monotonically increasing values than the lower threshold, the another upper and lower thresholds corresponding to a smaller degree of the monitored condition than the upper and lower thresholds; and

if the alarm signal violates the another upper threshold or the another lower threshold for a predetermined number of alarm signal values, triggers an output to a user indicative of the smaller degree of the monitored condition.

21. A predictive monitoring assembly for a battery assembly, comprising an analyzer configured to:

generate a temporal sequence of monotonically increasing values based on a series of monitored parameter samples, the monitored parameter samples obtained by monitoring a parameter associated with the battery assembly, the sequence including at least three values;

predict a fault condition of the battery assembly when the temporal sequence of monotonically increasing values indicates a sustained inclination in the monitored parameter toward one of an upper predetermined value or a lower predetermined value, the upper and lower predetermined values selected to correspond to a condition for which the battery is monitored; and

upon making the prediction, trigger an output to a user.

22. The predictive monitoring assembly of claim 21, wherein the analyzer derives an alarm signal from the temporal sequence, the alarm signal having a series of values based on an application of a Laplace test statistic to the temporal sequence values.

23. The predictive monitoring assembly of claim 21, wherein, to make the prediction, the analyzer:

derives an alarm signal that indicates a deviation in the temporal sequence of monotonically increasing values toward an upper threshold set to correspond to the upper predetermined value or a lower threshold set to correspond to the lower predetermined value;

compares the alarm signal to the upper threshold and the lower threshold; and

makes the prediction if a representation of the alarm signal violates the upper threshold or the lower threshold for a predetermined number of alarm signal values.

24. The predictive monitoring assembly of claim 21, further comprising a monitoring assembly that measures a parameter associated with the battery assembly.

25. The predictive monitoring assembly of claim 21, wherein the battery assembly is at least part of a lead acid battery system.

26. The predictive monitoring assembly of claim 21, wherein the battery assembly is monitored while in use.

27. The predictive monitoring assembly of claim 26, wherein the battery assembly is a primary power source for a load.

28. The predictive monitoring assembly of claim 26, wherein the battery assembly is a backup power source for a load.

29. The predictive monitoring assembly of claim 21, wherein the battery assembly is used in an electrical power transmission and distribution substation.

30. The predictive monitoring assembly of claim 21, wherein the monitored parameter is a function of at least one of voltage, current, impedance or temperature.

31. The predictive monitoring assembly of claim 23, wherein the analyzer is further configured to:

compare the alarm signal to another upper threshold that indicates less deviation in the temporal sequence of monotonically increasing values than the upper threshold and to another lower threshold that indicates less deviation in the temporal sequence of monotonically increasing values than the lower threshold, the another upper and lower thresholds corresponding to a smaller degree of the monitored condition than the upper and lower thresholds; and

if the alarm signal violates the another upper threshold or the another lower threshold for a predetermined number of alarm signal values, trigger an output to a user indicative of the smaller degree of the monitored condition.